|Publication number||US6938164 B1|
|Application number||US 09/721,398|
|Publication date||Aug 30, 2005|
|Filing date||Nov 22, 2000|
|Priority date||Nov 22, 2000|
|Also published as||EP1209563A2, EP1209563A3, EP1209563B1, US7543335, US7721341, US20050144476, US20050235166|
|Publication number||09721398, 721398, US 6938164 B1, US 6938164B1, US-B1-6938164, US6938164 B1, US6938164B1|
|Inventors||Paul England, Bryan Willman|
|Original Assignee||Microsoft Corporation|
|Export Citation||BiBTeX, EndNote, RefMan|
|Patent Citations (108), Non-Patent Citations (15), Referenced by (75), Classifications (11), Legal Events (5)|
|External Links: USPTO, USPTO Assignment, Espacenet|
This invention relates to secure code execution, and more particularly to a method and system for allowing code to be securely initialized in a computer.
Having people be able to trust computers has become an increasingly important goal. This trust generally focuses on the ability to trust the computer to use the information it stores or receives correctly. Exactly what this trust entails can vary based on the circumstances. For example, multimedia content providers would like to be able to trust computers to not improperly copy their content. By way of another example, users would like to be able to trust their computers to forward confidential financial information (e.g., bank account numbers) only to appropriate destinations (e.g., allow the information to be passed to their bank, but nowhere else). Unfortunately, given the generally open nature of most computers, a wide range of applications can be run on most current computers without the user's knowledge, and these applications can compromise this trust (e.g., forward the user's financial information to some other destination for malicious use).
To address these trust issues, different mechanisms have been proposed (and new mechanisms are being developed) that allow a computer or portions thereof to be trusted. Generally, these mechanisms entail some sort of authentication procedure where the computer can authenticate or certify that at least a portion of it (e.g., certain areas of memory, certain applications, etc.) are at least as trustworthy as they present themselves to be (e.g., that the computer or application actually is what it claims to be). In other words, these mechanisms prevent a malicious application from impersonating another application (or allowing a computer to impersonate another computer). Once such a mechanism can be established, the user or others (e.g., content providers) can make a judgment as to whether or not to accept a particular application as trustworthy (e.g., a multimedia content provider may accept a particular application as being trustworthy, once the computer can certify to the content provider's satisfaction that the particular application is the application it claims to be). However, installing such mechanisms on a computer can be difficult, as they require protection against a malicious application interfering with the mechanism (e.g., a malicious application impersonating the trusted mechanism).
One solution is to build a computer that includes a trustworthy mechanism for booting the computer. However, booting a computer typically involves using various pieces of code, often referred to as the basic input output system (BIOS) and potentially many option read only memories (ROMs) or BIOS extensions, that operate to load the operating system from some other storage device (typically a hard disk drive). Thus, a trustworthy mechanism for booting the computer would require that the BIOS be trustworthy, each of the option ROMs be trustworthy, and each of the BIOS extensions be trustworthy, before a trustworthy operating system can be loaded into the computer. Not only is it difficult to have each of these components trustworthy, but the BIOS, option ROMs, and BIOS extensions are frequently stored on devices that can be re-written in the computer (e.g., Flash memory), and thus the integrity thereof compromised. Therefore, building a computer that includes a trustworthy mechanism for booting the computer can be problematic.
The invention described below addresses these disadvantages, providing a method and system for allowing code to be securely initialized in a computer.
A method and system for allowing code to be securely initialized in a computer is described herein.
According to one aspect, a memory controller prevents CPUs and other I/O bus masters from accessing memory during a code (for example, OS, microkernel, or other trusted core) initialization process. The memory controller resets CPUs in the computer and allows a CPU to begin accessing memory at a particular location (identified to the CPU by the memory controller). Once an initialization process has been executed by that CPU, the code is operational and any other CPUs are allowed to access memory (after being reset), as are any other bus masters (subject to any controls imposed by the initiated code).
The present invention is illustrated by way of example and not limitation in the figures of the accompanying drawings. The same numbers are used throughout the figures to reference like components and/or features.
A method and system for allowing code (e.g., software instructions) to be securely initialized in a computer is described herein. This secure code initialization process is described primarily with reference to initializing a trusted core in a computer. However, the secure code initialization process can be used to securely initialize any of a wide variety of types of code, without regard for how trustworthy the code is. The mechanism described herein is of particular interest because no changes to the core-CPU, and few changes to the CPU supporting logic, are required.
As used herein, code being “trusted” refers to code that is immutable in nature and immutable in identity. Code that is trusted is immune to being tampered with by other parts (e.g. code) of the computer and it can be reliably and unambiguously identified. In other words, any other entity or component asking “who is this code” can be told “this is code xyz”, and can be assured both that the code is indeed code xyz (rather than some imposter) and that code xyz is unadulterated. Trust does not deal with any quality or usefulness aspects of the code—only immutability of nature and immutability of identity. The method and system described herein can be used to allow trusted code to know that it started execution in an orderly state, providing immutability of initialization.
Computer 100 includes one or more central processing units (CPUs) or processors 102, 104, a memory controller 106, an I/O controller 108, and system memory 110. CPUs 102 and 104 represent any of a wide variety of processors, such as x86-architecture (or compatible) processors, PowerPC-architecture (or compatible) processors, etc. Although multiple (n) CPUs 102, 104 are illustrated in
Memory controller 106 is illustrated as a separate component, but may alternatively be incorporated into another component or device, such as a bridge including an interrupt controller, accelerated graphics port (AGP), etc. Memory controller 106 communicates with CPUs 102, 104, I/O controller 108, and is memory 110. Memory 110 represents a system memory, such as a random access memory (RAM). I/O controller 108 operates as a bridge or interface between various I/O devices coupled to an I/O bus 114 and other components within computer 100. Any of a wide variety of conventional I/O devices can be coupled to I/O bus 114. In the illustrated example, a mass storage device controller 116 is coupled to bus 114 and a mass storage device (e.g., a hard disk) 116, a ROM 120 (including a BIOS 122 and option ROMs and/or BIOS extensions 124) is coupled to bus 114, and an adapter controller 126 is coupled to bus 114 and a network adapter 128 (e.g., network interface card (NIC) or modem).
Memory controller 106 includes a processor bus interface 130, a memory interface 132, an I/O interface 134, a control logic (also referred to as a processor or controller) 136, a digest 138, two reset vectors 140, 142, and initialization parameters 153. Processor bus interface 130 operates as an interface between memory controller 106 and processor bus 112, allowing memory controller 106 to receive commands and requests from CPUs 102, 104 as well as communicate responses and commands to CPUs 102, 104. Memory interface 132 operates as an interface to system memory 110, allowing controller 106 to read data stored in memory 110 and write data to memory 110. I/O interface 134 operates as an interface to I/O controller 108, allowing I/O components 118, 120, and 128 to communicate with memory controller 106 and vice versa via I/O controller 108. In the illustrated example, communications to and/or from a component internal to =controller 106 (e.g., control logic 136, reset vectors 140, 142, etc.) are passed through the appropriate interface.
Selected ones of I/O components coupled to I/O bus 114 may operate as bus masters. A bus master refers to a device (typically other than a CPU) that is capable of driving address bus signals and other bus control signals on a bus. Examples of bus masters are mass storage device controller 116 and network adapter controller 126. Bus masters can communicate with memory 110 via I/O controller 108 and 106, such as to perform direct memory access (DMA) transfers to and/or from memory 110, thereby alleviating a CPU 102, 104 from having to directly manage such transfers.
In the illustrated example of
Control logic 136 operates to ensure a trusted core can be loaded securely in computer 100, allowing accesses to memory 110 only under specific circumstances, as discussed in more detail below. Control logic 136 can be implemented in any of a variety of manners, such as a microcontroller executing microcode (stored in a nonvolatile secure location (e.g., within controller 106) accessible only to the microcontroller), a programmable logic device or other application specific integrated circuit (ASIC), etc.
Upon starting-up (e.g., powering-on or resetting) computer 100, the state of memory 110 is typically not guaranteed. Memory 110 is typically a dynamic RAM (DRAM) which requires periodic refreshing in order to maintain its state, so any state would have been lost when computer 100 was not powered. Memory 110 can be configured to be at a particular state (e.g., storing all zeroes or all ones) at power-on, or alternatively memory 110 may store random values (e.g., whatever values the individual memory cells stabilized to during power-on). Regardless, memory 110 stores no reliable data or instructions upon power-on. It should be noted that the contents of memory are more or less random when starting-up the entire computer system or just the memory. However, power-up and/or resetting of the CPU(s) do not typically alter the state of the memory (so the contents of memory are not necessarily random after simply resetting a CPU).
After computer 100 has been powered-on, one of CPUs 102, 104 will eventually begin executing instructions. Only one of CPUs 102, 104 may initially begin executing instructions, or alternatively one of CPUs 102, 104 may be selected to begin booting the computer. Any of a wide variety of arbitration mechanisms can be used to determine which of multiple CPUs is to begin booting the computer, such as a pre-configured default processor, a distributed arbitration scheme, an arbitration device or component (not shown) coupled to processor bus 112, etc.
Regardless of how many CPUs or which CPU begins execution of instructions, at least one CPU does begin executing instructions. As computer 100 has just been powered-on, no instructions are stored in the CPU's cache memory (or any other cache memory that may be present in computer 100). The CPU obtains instructions from BIOS 122 and begins executing those instructions. In one implementation, the CPU is configured to initially place a read request for a particular address (e.g., FFFFFFF016) on processor bus 112. Memory controller 106 responds with the address of a BIOS boot block (an address in BIOS 122) at which the CPU should begin executing instructions. This address returned by memory controller 112 is referred to as the reset vector, and in this (common) example, is provided by external logic. In an alternative implementation, the CPU simply begins fetching instructions directly from a particular location (for example, 0000000016), so in this case the reset vector is fixed.
The CPU then begins loading and executing instructions starting with this reset vector. The instructions loaded are from ROM 120, and can include instructions from BIOS 122 as well as one or more option ROMs and/or BIOS extensions 124. Some instructions from ROM 120 will result in instructions being loaded into memory 110 (either from ROM 120 or another device, such as mass storage device 118) that are then loaded by the CPU and executed.
Various components 144 of an operating system are thus loaded into memory 110, from ROM 120 and/or other devices (e.g., storage device 118). The components 144 can include portions of BIOS 122 as well as option ROMs and BIOS extensions 124. OS components 144 may also include selected components of a general purpose operating system, such as any of the Windows® family of operating systems available from Microsoft Corp. of Redmond, Wash.
Eventually, a trusted core is loaded into memory 110. This can be in response to, for example, instructions in BIOS 122, an option ROM or BIOS extension 124, an OS component 144, or another application (not shown). A trusted core refers to trustworthy code (e.g., software instructions) that controls at least a portion of the operation of computer 100, and knows how to protect itself (at least to some extent) on the particular hardware of computer 100. In protecting itself, the trusted core is also able to protect other applications running under the control of the trusted core. The amount of protection provided by the trusted core, as well as the manner in which such protection is provided by the trusted core, can vary among computers, computer (or CPU) architectures, and trusted cores. The trusted core may include all of the functionality of a typical operating system, or alternatively only selected functionality. The trusted core is typically cryptographically certified in some manner, allowing the integrity of the trusted core to be verified at a later time. For example, a cryptographic measure of the trusted core may be extracted from the trusted core and stored along with the trusted core on a mass storage device 118 (or on some remote device). When the trusted core is later loaded into memory 110, a component of computer 100 can extract the cryptographic measure of the trusted core in memory 110 for comparison to the stored cryptographic measure—if the two measures match then it can be safely assumed that the trusted core has not been altered (e.g., by a malicious user or program) while on device 118 or in memory 110.
In the illustrated example, the extracted cryptographic measure is stored in digest 138. Digest 138 refers to one or more registers (or other storage mechanisms) used to store a cryptographic measure of the trusted core as it is loaded. One such measure is a “cryptographic digest” that is calculated using a one-way hashing operation, such as SHA-1 (Secure Hash Algorithm 1). The cryptographic digest has the property that it is extremely difficult to find a second pre-image (in this case, a second trusted core) that when digested produces the same hash value. Hence the digest-register contains a value that can be considered to uniquely represent the trusted core in use. For a basic introduction of cryptography, the reader is directed to a text written by Bruce Schneier and entitled “Applied Cryptography: Protocols, Algorithms, and Source Code in C,” published by John Wiley & Sons with copyright 1994 (or second edition with copyright 1996).
Alternatively, measures other than a digest can be used. For example, the logic could require that the trusted core be accompanied by a digital certificate(s). The chipset could ensure that the certificate(s) matches the trusted core image, and then save the certificate(s), a public key(s) obtained from the certificate(s), or a digest(s) of the certificate(s) in the register.
Eventually, the computer is ready to load and initialize the component that will provides the “root of trust” for the running computer. A significant problem faced in previous computers is that all code that has executed beforehand can subvert the execution of the trusted core by programming the CPUs or other devices to misbehave. For instance, a bad OS-loader could arrange that a bus mastering controller should, after a brief delay, subvert the new component by writing untrusted code over it. The invention described herein removes all previously executing code from the trust chain.
The trusted core can operate in any of a wide variety of manners to make itself trustworthy. The manner in which the trusted core operates can vary, as can the amount of security or trustworthiness provided by the trusted core.
It should be noted that the invention described herein does not limit the nature of the code that will comprise the “root of trust.” It is anticipated that this code will take measures (such as programming the CPU memory controllers) to protect itself from code it might run, or devices it might program. However, the “root of trust” may be a full OS, a microkernel, a Hypervisor, or some smaller component that provides specific security services. Hereafter, we refer to such a component as the “trusted core.”
Alternatively, the operating system may be separated into a memory manager component that operates as trusted core 160 with the rest of the operating system operating as OS 162. The trusted core 160 then controls all page maps and is thus able to shield trusted agents executing in Ring 3 from other components (including OS 162). In this alternative, additional control also needs to be added (e.g., in memory controller 106 of
Trusted core data portion 152 is an area of memory 110 that is used to store data used by the trusted core. Platform trusted core portion 148 includes code that is specific to the particular hardware used in computer 100 (e.g., the specific type of CPU, the type of chipset, etc.). Common trusted core portion includes code that implements the operation and functionality of the trusted core but is not specific to the particular hardware used in computer 100. By separating the two portions 148 and 150, the specifics of how interaction with the hardware of computer 100 (e.g., how to set a timer) can be abstracted to portion 150 by portion 148, so the same common trusted core portion can be used on different computers. For example, the common trusted core code 150 can operate as if all timers were set in the same manner, while platform trusted core code 148 operates to convert the instructions from code 150 into commands that are actually understood by the underlying hardware of computer 100.
Trusted core 146 can be loaded into memory 110 from the same source or alternatively different sources. By way of example, platform trusted core portion 148 may be loaded into memory 110 from the chipset of computer 100 (the chipset refers to a set of one or more chips the provides the interfaces between the computer's subsystems, including the buses and electronics to allow the CPUs, memory, I/O devices, etc. to interact, and may include memory controller 106, I/O controller 108, and buses 112 and 114), while the common trusted core portion 150 may be loaded into memory 110 from mass storage device 118. By way of another example, one portion 148, 150, or 152 may be loaded into memory 110 from network adapter 128 (e.g., from a remote server), or the portions may all be loaded from network adapter 128 but from different sources (e.g., different remote servers). Additionally, different parts of one or more of portions 148, 150, and 152 may be loaded from different sources (e.g., part of portion 150 may be loaded into memory 110 from mass storage device 118 and another part of portion 150 may be loaded from a remote server via network adapter 128).
Additionally, each portion may be generated by combining various overlaid parts from one or more sources. For example, a first part could be read from one source (e.g., a local disk) and a second part read from another source (e.g., a remote server over the network), and then the two parts could be combined (e.g., exclusive-OR the bits of each part) to generate the portion.
It should be noted that the trusted core 146 need not be stored in a secure location and/or loaded into memory 110 in a secure manner. The trusted core 146 may be stored on any publicly-accessible (or semi-publicly accessible) area, such as mass storage device 118, a remote server accessed via network adapter 128, etc. Rather, a cryptographic measure of trusted core 146 can be verified after loading into memory 110 (and after it is protected so that no further modifications by untrusted code are possible). The cryptographic measure can be generated in any of a wide variety of conventional manners, and is designed so that any change in the trusted core is reflected in the cryptographic measure. For example, if a malicious user were to attempt to alter the trusted core by deleting or modifying an instruction(s), or adding a new instruction(s), in an attempt to subvert the security of the trusted core, that alteration would be reflected in the cryptographic measure (so any subsequently generated cryptographic measure would not match a previously generated cryptographic measure due to the change in the instruction(s)). In one implementation, the cryptographic measure is obtained by generating a digest value of the trusted core, such as in accordance with the well-known SHA, SHA-1, MD5 (Message Digest 5), etc. algorithms.
This cryptographic measure is generated during the secure initialization process described herein, and is made subsequently available to challengers (e.g., a bank, IT administrator, smart card, etc.). Interpreting the digest can be done in several ways. It could be a “well known” value, such as a company publishing the expected value of the its trusted core on its web site. Alternatively, the company may generate a certificate for each trusted core it writes, and ship it with the trusted core (or make it available on a web site). This allows challengers to see the digest and ask for (or look for) a certificate that “names” the digest. It should be noted that this certificate should not be in the core itself because the certificate names the digest.
It should be noted that the initialization process described herein does not limit what a user may try to run as the “trusted core”—the user can try to run virtually any set of instructions as the trusted core. However, if the trusted core is tampered with, then the instructions attempting to run as the “trusted core” (the tampered with trusted core) will not be trusted by other parties because the tampered with trusted core will have a different cryptographic measure and will not have access to keys that are accessible to the non-tampered with trusted core.
It should also be noted that the initialization process described herein naturally and easily allows a computer to run different trusted cores serially, at arbitrary times, without re-booting the system. For example, the user can begin executing trusted core x, then switch to executing trusted core y (which may be a different version of the trusted core x, or an entirely different trusted core), and then switch to trusted core z (or back to trusted core x), etc.
Memory controller 106 operates to ensure that trusted core 146, once loaded into memory 110, can be initialized and begin execution in such a manner that any previous state of computer 100 does not compromise the integrity of the initialization process. It should be noted memory controller 106 operates to provide a secure environment in which the trusted core can be initialized. However, once the initialization process is completed, trusted core 146 is responsible for its own security. Additionally, memory controller 106 does not generate any guarantees or assurances regarding the trustworthiness or security provided by trusted core 146.
Initialization of trusted core 146 refers to beginning execution of trusted core 146. This initialization typically occurs after trusted core 146 has been loaded in its entirety into memory 110, although alternative implementations may allow for some parts of code 146 to be loaded into memory after the initialization process occurs. The initialization process is performed by a set of trusted core initialization instructions that are part of trusted core 146 (and maybe part of one or more of portions 148, 150, and 152). The exact acts performed by the initialization process can vary based on the specific manner in which the trusted core operates as well as the specific hardware of computer 100 (e.g., the CPU architecture), but generally include establishing the protection provided by the trusted core (e.g., setting up a trusted space) and initializing CPUs to use this protection. One of the things the “initialization code” must do is provide a “reset vector” which is left in place, and is always ready to deal with, any processor that is reset, initiated, hot-plug added, etc. By doing this, the trusted core ensures its continued integrity in the face of profound changes in the CPU configuration of the machine (e.g. having new ones added on the fly.)
Initialization of trusted core 146 begins in response to a “Trusted Core Initialization” command issued by one of CPUs 102, 104. The Trusted Core Initialization command is commonly issued during booting of computer 100, although it may occur any time during or after computer 100 has been fully booted. The Trusted Core Initialization command is designed to trigger initialization of the trusted core, and can be implemented in any of a wide variety of manners. For example, the command could be a write to a particular port, a write to a particular address, etc., however, the port or address should have the property that the trusted core can protect the port or address using whatever protection means that hardware provides after the core has gained control. This is to ensure that once a trusted core is running, it is not vulnerable to the simple denial of service attack that any untrusted code can remove it from memory or replace it. Alternatively, if such protection is not provided (e.g., the Trusted Core Initialization command has to be invoked by some unprotectable port) then it may be a “run once” operation which refuses to perform its function again until the system has been rebooted. However, such a machine loses the ability to stop and start trusted cores or to use more than one trusted core per boot.
In one implementation, the Trusted Core Initialization command includes three parameters: start, length of code, and length of memory. The start parameter refers to the location in memory 110 where trusted core 146 begins (e.g., the memory address of the first instruction of platform trusted code portion 148). Alternatively, especially in embodiments where trusted core 146 is not located in physically contiguous memory, a memory descriptor list (MDL) can be identified as the start parameter, identifying to memory controller 106 which memory pages include the trusted core. The length of code parameter refers to the length (e.g., the number of bytes) of the code of trusted core 146 (e.g., the combination of portions 148 and 150, but excluding trusted core data portion 152). The length of memory parameter refers to the length (e.g., the number of bytes) of trusted core (e.g., the combination of all three portions 148, 150, and 152). These parameters are passed to memory controller 106 either prior to (or as part of) the Trusted Core Initialization command. In one implementation, these parameters are loaded by one of CPUs 102, 104 into parameter registers 153 of memory controller 106 prior to issuance of the Trusted Core Initialization command.
Upon receipt of the Trusted Core Initialization command, memory controller 106 protects memory 110 from all CPUs 102, 104 and any other bus masters in computer 110. Given that CPUs 102, 104, as well as any other bus masters in computer 100, cannot directly access memory 110, but rather have to access memory 110 via memory controller 106, memory controller 106 can readily protect memory 110 from the CPUs and bus masters. By protecting memory 110, memory controller 106 ensures that other code (e.g., being executed by a CPU or controlling another bus master) does not interfere with the trusted core initialization process. In protecting memory 110, memory controller 106 initially allows the only access to memory 110 to be the memory refresh signals from a memory refresh controller (not shown) that allow DRAM to be refreshed and maintain its state.
On a machine with factored memory—that is, multiple blocks of memory with multiple memory controllers—multiple different approaches may be used. In one approach, the multiple memory controllers coordinate with each other to apply the Trusted Core Initialization operator across all memories. In another approach, the memory-controller/memory selected to do the Trusted Core Initialization does all of it (so the whole trusted core is loaded into one “bank”) and other memory controllers ignore it. Since all of the trusted core is in a single protected memory bank, the activities of the others do not matter. The selected memory controller still forces the behavior of all CPUs, so the platform hardware is set up to allow it to do that.
Control logic 136 in memory controller 106 also sets a trusted core initialization bit 154 within a secure portion 156 of memory controller 106. This secure portion 156 is secure in that only control logic 136 is allowed to access the registers and/or memory within secure portion 156. Secure portion 156 may not be made visible to other components in computer 100 external to memory controller 106, or alternatively control logic 136 may simply ignore (or reject) any commands received at memory controller 106 attempting to access any register or memory within secure portion 156. The trusted core initialization bit 154 allows control logic 136 to know whether the trusted core initialization process is in-process, and the use of bit 154 is discussed in more detail below.
Memory controller 106 can protect memory 110 in a variety of different manners. In one implementation, memory controller 106 prevents CPUs 102, 104 from issuing read or write requests on processor bus 112, thereby preventing CPUs 102, 104 from accessing memory 110. The manner in which CPUs 102, 104 are kept off the bus varies based on the architecture of CPUs 102, 104 and processor bus 112 (e.g., in certain Intel-architecture implementations, CPUs can be prevented form issuing read and write requests on the bus by asserting a BNR# (Block Next Request) signal on processor bus 112, or by issuing a halt (e.g., HLT) command to the CPUs which halts the operation of each CPU until it is reset). CPUs 102, 104 will eventually be allowed to again issue read and write requests on processor bus 112, so control logic 136 also compares all requests that memory controller 106 receives from a CPU 102, 104 to the address range of trusted core 146 in memory 110—any request that memory controller 106 receives from a CPU 102, 104 during this trusted core initialization process (e.g., with bit 154 set) that is within this address range is accepted and responded to in a conventional manner, while any such request that is not within this address range or that is from a component other than a CPU 102, 104 is rejected (e.g., memory controller 106 may refuse to acknowledge the request, may deny the request, etc.). Alternatively, requests received that are not within this address range may be allowed, but with the understanding that such areas of memory are not protected.
Memory controller 106 also prevents other bus masters on I/O bus 114 from accessing memory 110. Additionally, memory controller 106 can prevent other bus masters on I/O bus 114 from affecting cache memories (e.g., within a CPU or on bus 112), such as by not issuing any such transactions onto bus 112. For example, memory controller 106 may have I/O controller 108 assert a signal on I/O bus 114 that prevents any of the bus masters from issuing read and write requests on bus 114. By way of another example, memory controller 106 may simply ignore any memory access requests it receives from I/O controller 108, such as by having I/O interface 134 deny such requests, refuse to accept or acknowledge such requests, etc. Although preventing access to memory 110 can be implemented in similar fashions for both CPU accesses and I/O bus master accesses, it should be noted that no I/O bus master accesses are allowed during the initialization process even though CPU access are allowed.
Memory controller 106 may immediately protect memory 110 upon receipt of the Trusted Core Initialization command, or alternatively wait for a period of time or for some other action to occur before protecting memory 110. For example, a particular bus master may be in the process of a long DMA transfer at the time the Trusted Core Initialization command is received, and memory controller 106 may opt to let the DMA transfer finish. Although the protection of memory 110 may not be immediate, the trusted core initialization process does not proceed until memory 110 is protected.
Once memory 110 is protected, control logic 136 generates a digest value based on trusted core 146. Alternatively, another component (not shown), such as a dedicated cryptographic processor, may be temporarily allowed access to memory 110 in order to generate the digest value (or the digest could be held in the cryptographic-processor, as long as the cryptographic-processor works intimately with the memory controller to perform the digest operation only when the CPU is performing the initialization process (and cannot be fooled by untrusted software making this request of the cryptographic-processor)). This access, however, may be limited, such as allowing the cryptographic processor to only read from memory 110. This digest is generated using the same cryptographic mechanism as discussed above. Control logic 136 then stores the generated digest value as digest 138 in secure portion 156 of memory controller 106. Digest 138 can then be used for subsequent certification and/or authentication purposes by trusted core 146.
Control logic 136 then maps the reset vector to a new reset vector, referred to as the initialization vector, so that memory controller 106 responds to the next read request for a particular address (e.g., FFFFFFF016) on processor bus 112 with the initialization vector. In situations where the CPU always starts executing from a known location (rather than using the indirection supplied by the reset-vector fetch procedure), control logic 136 arranges to issue an instruction(s) as the first one or few instructions that are fetched from the memory subsystem that causes the CPU to begin executing instructions at the start of the trusted core (e.g., a JUMP operation), or alternatively re-map the trusted core to appear at the fixed processor reset vector. This initialization vector is the first address of trusted core 146. In one implementation, this mapping is accomplished by control logic 136 writing the start location of trusted core 146 (e.g., received as a parameter of the Trusted Core Initialization command) as reset vector 140. Whenever memory controller 106 receives a read request for the particular address (e.g., FFFFFFF016) on processor bus 112, control logic 136 checks bit 154 of secure portion 156. If the bit is set, then the mapped reset vector 140 is returned to the requesting CPU as the reset vector. However, if the bit is not set, then the initial BIOS boot block address is returned to the requesting CPU.
Control logic 136 then asserts a reset signal to a CPU on processor bus 112. In the situation where multiple CPUs are on processor bus 112, the reset signal is asserted to all of the CPUs (although alternatively the reset signal may be directed to a single CPU). The manner in which the reset signal is asserted (and the nature of the reset signal itself) can vary based on the architecture of CPUs 102, 104 and processor bus 112 (e.g., in certain Intel-architecture implementations, all CPUs on processor bus 112 can be reset by asserting a RESET# signal on processor bus 112).
Alternatively, other mechanisms can be used to reset the CPUs on processor bus 112. The goal of resetting a CPU is to clear the CPU of its state (e.g., all instructions and data that may reside within registers, caches, buffers, etc. of the CPU) so that that prior state cannot compromise the execution integrity of the trusted core initialization process. For example, so that instructions of a malicious application that are stored in a CPU's cache cannot compromise the integrity of the trusted core initialization process.
In addition to resetting the CPUs on processor bus 112, any other cache memory or other device on processor bus 112 that includes memory is reset. For example, an additional L3 cache (not shown) may also be coupled to processor bus 112, in which case control logic 136 would reset that L3 cache as well as the CPUs. These additional cache memories or other devices on processor bus 112 can be reset in any of a variety of manners, such as writing random values to the memories, writing a string of zeroes or ones to the memories, removing power from the memories (e.g., temporarily not refreshing DRAMs), issuing specific bus transactions to invalidate entire caches or specific cache-line entries, etc.
After resetting CPUs 102, 104 (and any other caches or devices on processor bus 112 (except memory controller 106)), memory controller 106 allows one of the CPUs on processor bus 112 to access memory 110. In the illustrated example, any of the CPUs 102, 104 can be initially allowed to access memory 110 (although only one is allowed). Memory controller 106 can determine which of multiple processors to allow to access memory 110 in any of a variety of manners, such as some pre-determined setting, a random determination, allowing some external (or CPU-internal) bus arbitration mechanism to determine, etc.
For the purposes of discussion, assume that memory controller 106 determines to allow CPU 102 to access memory 110. Memory controller 106 then allows CPU 102 to access processor bus 112 (all other CPUs 104 are still prevented from accessing processor bus 112). Once a CPU is reset, the CPU will have no instructions (or at least no instructions or data that the CPU will read) in any of its caches or buffers. CPU 102 will thus request instructions to execute by asserting the particular address (e.g., FFFFFFF016) on processor bus 112. Memory controller 106 receives this request, control logic 136 determines that trusted core initialization process bit 154 is set, so memory controller 106 returns reset vector 140 to CPU 102. Also upon receipt of the particular address (e.g., FFFFFFF016) control logic sets an additional per-processor vector bit 158. This bit 158 allows control logic 136 to know, in the event another particular address (e.g., FFFFFFF016) is received, that reset vector 140 has already been returned to another processor. The use of bit 158 is discussed in more detail below.
Upon receipt of reset vector 140, CPU 102 issues a read request on processor bus 112 for the memory address in reset vector 140, which is the initial instruction in trusted core 146 (and the initial instruction in the trusted core initialization process). CPU 102 thus begins to execute the trusted core code 146. Since CPU 102 was reset, no instructions (e.g., malicious code) previously loaded into CPU 102 can affect its execution of trusted core 146. Additionally, since all other bus masters are prevented from accessing memory 110, they cannot alter the code 146 being fetched and executed by CPU 102 from memory 110.
CPU 102 continues to fetch and execute trusted core code 146, which initializes the trusted core in computer 100. As discussed above, the exact manner in which the trusted core is initialized can vary based on the architecture of computer 100, CPUs 102, 104, the type of trusted core being implemented, etc. As part of this initialization process, if trusted core 146 supports multiple processors, an additional CPU start vector is loaded as reset vector 142. This additional CPU start vector identifies the memory address in code 146 where additional instructions are to be fetched in the event additional processors begin executing instructions. This second reset vector can be used, for example, to have additional processors begin executing instructions elsewhere in the trusted core than the beginning of trusted core 146 (e.g., other than the beginning of the trusted core initialization part of code 146). However, since no other CPU is allowed to run until the “End of Initialization” the vector at 142 won't be used until after “End of Initialization”.
Eventually the trusted core initialization process completes, and an “End of Initialization” command is issued by CPU 102. The End of Initialization command can be asserted in any of a variety of manners, analogous to the Trusted Core Initialization command discussed above. Upon receipt of the End of Initialization command, memory controller 106 allows all remaining CPUs 104 on processor bus 112 to access memory 110. Each of these CPUs 104, having been reset, will have no instructions (or at least no valid instructions) in any of their caches or buffers. Each CPU 104 will thus request instructions to execute by asserting the particular address (e.g., FFFFFFF016) on processor bus 112. Memory controller 106 receives these requests, control logic 136 determines that additional processor vector bit 158 is set, so memory controller 106 returns reset vector 142 to each of CPUs 104.
Control logic 136 then clears the trusted core initialization bit 154, as the trusted core initialization process is complete. Memory controller 106 also ceases preventing other bus masters 110 from accessing memory 110. Thus, any additional protection against malicious code from another bus master is to be prevented by the execution of the trusted core 146, not memory controller 106.
Each CPU 104, upon receipt of reset vector 142, issues a read request on processor bus 112 for the memory address in reset vector 142, which is an instruction in trusted core 146. Each CPU 104 thus begins to execute instructions in the trusted core code 146. Since each CPU 104 was reset, no instructions (e.g., malicious code) previously loaded into the CPU 104 can affect its execution of trusted core 146.
In one implementation, one or more of CPU 102, 104 may need microcode loaded into it in order to function properly. This microcode may, for example, fix errors or other bugs in the CPU, expose a different instruction set than the native instruction set of the CPU, etc. This microcode can be included in trusted core 146 (e.g., in trusted core data portion 152) and instructions included in the trusted core initialization process to load the microcode into the appropriate CPU(s). Additionally, the authenticity of the microcode may be verified, such as by re-computing a digest and comparing it to a certified digest within the microcode (e.g., placed there by the author or distributor of the microcode) to ensure that the microcode has not been altered since being signed by the distributor of the microcode.
In one implementation, memory controller 106 provides the following assurances in response to a Trusted Core Initialization command:
In addition, memory controller 106 may also operate to intercept and disallow any interrupts. For example, memory controller 106 may include, or alternatively be in communication with, the interrupt controller(s) for computer 100. Any interrupt received during the trusted core initialization process can be disallowed. Alternatively, memory controller 106 may not concern itself with interrupts (e.g., because the CPUs 102, 104 are reset, and limited by memory controller 106 and trusted core 146 in their ability to obtain instructions).
Additionally, in certain embodiments, the trusted core initialization process establishes certain regions of memory (either real and/or virtual regions) that are within the protected or curtained space of the trusted core. During operation of the trusted core, the trusted core operates to ensure that only trusted applications executing within that protected space can access other memory addresses within that protected space. In these embodiments, all controls of memory controller 106 (e.g., addressable portions of controller 106, such as parameter registers 153) are included within the protected space. Additionally, the mechanism via which the Trusted Core Initialization command is issued (e.g., a particular register address, a particular port, etc.) is also included in the protected space, preventing untrustworthy code from re-issuing the Trusted Core Initialization command.
In certain implementations, CPUs 102, 104 support two modes of operation: a real mode and a protected mode. Real mode refers to offering the programming environment of a predecessor CPU (e.g., the Intel 8086 processor), while protected mode makes all instructions and architectural features of the CPU available. In these implementations, CPUs 102, 104 are initialized into real mode at power-on (and in response to a RESET# signal). The trusted core initialization process begins, for each CPU, by transitioning that CPU into protected mode and then turning memory paging on (memory paging refers to virtual memory, where “pages” of memory can be referenced by the CPU and swapped between memory 110 and another storage device (e.g., device 118) as needed). Care should thus be taken in the trusted core initialization process that for additional CPUs (after the initial CPU) that are reset and permitted to access memory, they be able to run in real mode for long enough to be transitioned to protected mode and paging turned on. This can be accomplished, for example, by care in selection of the additional processor start vector (reset vector 142).
Note also that if code external to the trusted core arranges a forced processor reset (this is often possible programmatically, and is typically possible on PC-class machines), then the processor will start up by executing the trusted core, and not code that can potentially subvert a running core. In this case, the core may choose to treat this as an exception condition, and destroy all existing internal state (clear memory), or can take action to continue execution (for example, reload the CPU protection profile that was in effect before the processor reset).
It should be noted that the trusted core initialization process does not require any additional bus transactions on processor bus 112, no additional pins on CPUs 102, 104, or other such modifications to the CPUs 102, 104. Rather, the initialization process is implemented in the memory controller (e.g., in the chipset), thereby preventing the need for any modifications to the CPUs.
Although discussed herein as primarily performed during booting of a computer, the trusted core initialization process can be performed at different times. For example, the trusted core may not be loaded into memory 110 and the Trusted Core Initialize command issued until an application that requires the trusted core to operate is to be executed (e.g., the user is going to download multimedia content that needs to be protected). Additionally, the trusted core initialization process can be repeated multiple times. By way of example, the trusted core code may have a “Terminate Trusted Core” command that terminates execution of the trusted core. However, the trusted core could again be loaded into memory 110 (if it were actually removed as part of the execution termination) at a later time, and the trusted core initialization process repeated for the trusted core code to begin executing again. And, multiple different trusted cores can be run, one at a time, one after the other.
Additionally, it should be noted that after the trusted core initialization process has completed and other CPUs are allowed to access memory, these additional CPUs need not have been present in the computer when the trusted core initialization process began. By way of example, the computer may support the ability to “hot-plug” CPUs (add CPUs to the computer while the computer is running). Any such newly added CPUs are allowed to access memory, starting execution of instructions at the additional processor reset vector.
In the example of
Initially, the normal boot process for the computer is started using the not necessarily secure BIOS (act 252). Eventually, the code is loaded into memory (act 254) and an Initialize Code command received (act 256). In response to the Initialize Code command, memory is protected from all processors and bus masters (act 258). A cryptographic measure of the code in memory is then extracted and stored in a secure location (act 260). The processor reset vector is then mapped to an initialization vector (act 262), such as the start of the code in the memory.
Once the processor reset vector is mapped to the initialization vector, all the processors are reset (act 264) and one of the processors is allowed to access memory (act 266). The process then waits for the processor that is allowed to access memory to execute a code initialization process from the memory (act 268). The end of the code initialization process is signaled, for example, by a command issued from the processor executing code initialization process. Once the code initialization process is finished, the processor reset vector is re-mapped to an additional processor start vector (act 270). Any other processors in the computer are then allowed to access memory (act 272), as are any other bus masters in the computer (act 274).
In the discussion herein, embodiments of the invention are described in the general context of processor-executable instructions, such as program modules or code, being executed by one or more conventional processors or controllers (e.g., control logic 136 of
Alternatively, embodiments of the invention can be implemented in hardware or a combination of hardware, software, and/or firmware. For example, all or part of the invention can be implemented in one or more application specific integrated circuits (ASICs) or programmable logic devices (PLDs). Computing devices (such as device 100 of
For purposes of illustration, programs and other executable program components such as the operating system are illustrated herein as discrete blocks, although it is recognized that such programs and components reside at various times in different storage components of the computer, and are executed by the data processor(s) of the computer.
Although the description above uses language that is specific to structural features and/or methodological acts, it is to be understood that the invention defined in the appended claims is not limited to the specific features or acts described. Rather, the specific features and acts are disclosed as exemplary forms of implementing the invention.
|Cited Patent||Filing date||Publication date||Applicant||Title|
|US4827508||Oct 14, 1986||May 2, 1989||Personal Library Software, Inc.||Database usage metering and protection system and method|
|US4969189||Jun 19, 1989||Nov 6, 1990||Nippon Telegraph & Telephone Corporation||Authentication system and apparatus therefor|
|US4977594||Feb 16, 1989||Dec 11, 1990||Electronic Publishing Resources, Inc.||Database usage metering and protection system and method|
|US5023907||Sep 30, 1988||Jun 11, 1991||Apollo Computer, Inc.||Network license server|
|US5050213||Aug 6, 1990||Sep 17, 1991||Electronic Publishing Resources, Inc.||Database usage metering and protection system and method|
|US5140634||Oct 9, 1991||Aug 18, 1992||U.S Philips Corporation||Method and apparatus for authenticating accreditations and for authenticating and signing messages|
|US5175829 *||Oct 25, 1988||Dec 29, 1992||Hewlett-Packard Company||Method and apparatus for bus lock during atomic computer operations|
|US5276311||Jul 1, 1992||Jan 4, 1994||Hartmut Hennige||Method and device for simplifying the use of a plurality of credit cards, or the like|
|US5335334||Aug 29, 1991||Aug 2, 1994||Hitachi, Ltd.||Data processing apparatus having a real memory region with a corresponding fixed memory protection key value and method for allocating memories therefor|
|US5349643 *||May 10, 1993||Sep 20, 1994||International Business Machines Corporation||System and method for secure initial program load for diskless workstations|
|US5408647||Oct 2, 1992||Apr 18, 1995||Compaq Computer Corporation||Automatic logical CPU assignment of physical CPUs|
|US5410598||Sep 27, 1994||Apr 25, 1995||Electronic Publishing Resources, Inc.||Database usage metering and protection system and method|
|US5473690||Jan 16, 1992||Dec 5, 1995||Gemplus Card International||Secured method for loading a plurality of applications into a microprocessor memory card|
|US5473692||Sep 7, 1994||Dec 5, 1995||Intel Corporation||Roving software license for a hardware agent|
|US5491827||Jan 14, 1994||Feb 13, 1996||Bull Hn Information Systems Inc.||Secure application card for sharing application data and procedures among a plurality of microprocessors|
|US5511184||Oct 22, 1993||Apr 23, 1996||Acer Incorporated||Method and apparatus for protecting a computer system from computer viruses|
|US5544246||Sep 17, 1993||Aug 6, 1996||At&T Corp.||Smartcard adapted for a plurality of service providers and for remote installation of same|
|US5557518||Apr 28, 1994||Sep 17, 1996||Citibank, N.A.||Trusted agents for open electronic commerce|
|US5654746||Dec 1, 1994||Aug 5, 1997||Scientific-Atlanta, Inc.||Secure authorization and control method and apparatus for a game delivery service|
|US5664016||Oct 17, 1995||Sep 2, 1997||Northern Telecom Limited||Method of building fast MACS from hash functions|
|US5671280||Aug 30, 1995||Sep 23, 1997||Citibank, N.A.||System and method for commercial payments using trusted agents|
|US5721781||Sep 13, 1995||Feb 24, 1998||Microsoft Corporation||Authentication system and method for smart card transactions|
|US5724527||Dec 28, 1995||Mar 3, 1998||Intel Corporation||Fault-tolerant boot strap mechanism for a multiprocessor system|
|US5745886||Jun 7, 1995||Apr 28, 1998||Citibank, N.A.||Trusted agents for open distribution of electronic money|
|US5757919||Dec 12, 1996||May 26, 1998||Intel Corporation||Cryptographically protected paging subsystem|
|US5778070 *||Jun 28, 1996||Jul 7, 1998||Intel Corporation||Method and apparatus for protecting flash memory|
|US5787427||Jan 3, 1996||Jul 28, 1998||International Business Machines Corporation||Information handling system, method, and article of manufacture for efficient object security processing by grouping objects sharing common control access policies|
|US5796824||Feb 20, 1996||Aug 18, 1998||Fujitsu Limited||Storage medium for preventing an irregular use by a third party|
|US5812662||Dec 18, 1995||Sep 22, 1998||United Microelectronics Corporation||Method and apparatus to protect computer software|
|US5812980||Feb 16, 1995||Sep 22, 1998||Sega Enterprises, Ltd.||Program operating apparatus|
|US5841869||Aug 23, 1996||Nov 24, 1998||Cheyenne Property Trust||Method and apparatus for trusted processing|
|US5867655 *||Aug 19, 1997||Feb 2, 1999||Packard Bell Nec||Method to store privileged data within the primary CPU memory space|
|US5872847||Jul 30, 1996||Feb 16, 1999||Itt Industries, Inc.||Using trusted associations to establish trust in a computer network|
|US5892900||Aug 30, 1996||Apr 6, 1999||Intertrust Technologies Corp.||Systems and methods for secure transaction management and electronic rights protection|
|US5892902||Sep 5, 1996||Apr 6, 1999||Clark; Paul C.||Intelligent token protected system with network authentication|
|US5892904||Dec 6, 1996||Apr 6, 1999||Microsoft Corporation||Code certification for network transmission|
|US5910987||Dec 4, 1996||Jun 8, 1999||Intertrust Technologies Corp.||Systems and methods for secure transaction management and electronic rights protection|
|US5915019||Jan 8, 1997||Jun 22, 1999||Intertrust Technologies Corp.||Systems and methods for secure transaction management and electronic rights protection|
|US5917912||Jan 8, 1997||Jun 29, 1999||Intertrust Technologies Corporation||System and methods for secure transaction management and electronic rights protection|
|US5919257||Aug 8, 1997||Jul 6, 1999||Novell, Inc.||Networked workstation intrusion detection system|
|US5920861||Feb 25, 1997||Jul 6, 1999||Intertrust Technologies Corp.||Techniques for defining using and manipulating rights management data structures|
|US5933498||Nov 5, 1997||Aug 3, 1999||Mrj, Inc.||System for controlling access and distribution of digital property|
|US5940504||Jun 29, 1992||Aug 17, 1999||Infologic Software, Inc.||Licensing management system and method in which datagrams including an address of a licensee and indicative of use of a licensed product are sent from the licensee's site|
|US5943422||Aug 12, 1996||Aug 24, 1999||Intertrust Technologies Corp.||Steganographic techniques for securely delivering electronic digital rights management control information over insecure communication channels|
|US5944821||Jul 11, 1996||Aug 31, 1999||Compaq Computer Corporation||Secure software registration and integrity assessment in a computer system|
|US5949876||Jan 8, 1997||Sep 7, 1999||Intertrust Technologies Corporation||Systems and methods for secure transaction management and electronic rights protection|
|US5953502||Feb 13, 1997||Sep 14, 1999||Helbig, Sr.; Walter A||Method and apparatus for enhancing computer system security|
|US5958050||Dec 26, 1996||Sep 28, 1999||Electric Communities||Trusted delegation system|
|US5963980||Dec 6, 1994||Oct 5, 1999||Gemplus Card International||Microprocessor-based memory card that limits memory accesses by application programs and method of operation|
|US5982891||Nov 4, 1997||Nov 9, 1999||Intertrust Technologies Corp.||Systems and methods for secure transaction management and electronic rights protection|
|US5991399||Dec 18, 1997||Nov 23, 1999||Intel Corporation||Method for securely distributing a conditional use private key to a trusted entity on a remote system|
|US5991876||Apr 1, 1996||Nov 23, 1999||Copyright Clearance Center, Inc.||Electronic rights management and authorization system|
|US6006332||Oct 21, 1997||Dec 21, 1999||Case Western Reserve University||Rights management system for digital media|
|US6009274||Jun 24, 1997||Dec 28, 1999||3Com Corporation||Method and apparatus for automatically updating software components on end systems over a network|
|US6009401||Apr 6, 1998||Dec 28, 1999||Preview Systems, Inc.||Relicensing of electronically purchased software|
|US6026166||Oct 20, 1997||Feb 15, 2000||Cryptoworx Corporation||Digitally certifying a user identity and a computer system in combination|
|US6032257||Aug 29, 1997||Feb 29, 2000||Compaq Computer Corporation||Hardware theft-protection architecture|
|US6038551||Mar 11, 1996||Mar 14, 2000||Microsoft Corporation||System and method for configuring and managing resources on a multi-purpose integrated circuit card using a personal computer|
|US6073124||Jul 15, 1997||Jun 6, 2000||Shopnow.Com Inc.||Method and system for securely incorporating electronic information into an online purchasing application|
|US6092189||Apr 30, 1998||Jul 18, 2000||Compaq Computer Corporation||Channel configuration program server architecture|
|US6105137||Jul 2, 1998||Aug 15, 2000||Intel Corporation||Method and apparatus for integrity verification, authentication, and secure linkage of software modules|
|US6112181||Nov 6, 1997||Aug 29, 2000||Intertrust Technologies Corporation||Systems and methods for matching, selecting, narrowcasting, and/or classifying based on rights management and/or other information|
|US6118873||Apr 24, 1998||Sep 12, 2000||International Business Machines Corporation||System for encrypting broadcast programs in the presence of compromised receiver devices|
|US6138119||Apr 27, 1999||Oct 24, 2000||Intertrust Technologies Corp.||Techniques for defining, using and manipulating rights management data structures|
|US6148387||Jun 18, 1999||Nov 14, 2000||Phoenix Technologies, Ltd.||System and method for securely utilizing basic input and output system (BIOS) services|
|US6148402||Apr 1, 1998||Nov 14, 2000||Hewlett-Packard Company||Apparatus and method for remotely executing commands using distributed computing environment remote procedure calls|
|US6157721||Aug 12, 1996||Dec 5, 2000||Intertrust Technologies Corp.||Systems and methods using cryptography to protect secure computing environments|
|US6175917||Apr 23, 1998||Jan 16, 2001||Vpnet Technologies, Inc.||Method and apparatus for swapping a computer operating system|
|US6185678||Oct 2, 1998||Feb 6, 2001||Trustees Of The University Of Pennsylvania||Secure and reliable bootstrap architecture|
|US6185683||Dec 28, 1998||Feb 6, 2001||Intertrust Technologies Corp.||Trusted and secure techniques, systems and methods for item delivery and execution|
|US6189100||Jun 30, 1998||Feb 13, 2001||Microsoft Corporation||Ensuring the integrity of remote boot client data|
|US6192473||Dec 24, 1996||Feb 20, 2001||Pitney Bowes Inc.||System and method for mutual authentication and secure communications between a postage security device and a meter server|
|US6212636||May 1, 1997||Apr 3, 2001||Itt Manufacturing Enterprises||Method for establishing trust in a computer network via association|
|US6223284||Apr 30, 1998||Apr 24, 2001||Compaq Computer Corporation||Method and apparatus for remote ROM flashing and security management for a computer system|
|US6229894||Jul 14, 1997||May 8, 2001||Entrust Technologies, Ltd.||Method and apparatus for access to user-specific encryption information|
|US6230285||Sep 8, 1998||May 8, 2001||Symantec Corporation||Boot failure recovery|
|US6237786||Jun 17, 1999||May 29, 2001||Intertrust Technologies Corp.||Systems and methods for secure transaction management and electronic rights protection|
|US6240185||Feb 10, 1999||May 29, 2001||Intertrust Technologies Corporation||Steganographic techniques for securely delivering electronic digital rights management control information over insecure communication channels|
|US6253193||Dec 9, 1998||Jun 26, 2001||Intertrust Technologies Corporation||Systems and methods for the secure transaction management and electronic rights protection|
|US6263431||Dec 31, 1998||Jul 17, 2001||Intle Corporation||Operating system bootstrap security mechanism|
|US6272629||Dec 29, 1998||Aug 7, 2001||Intel Corporation||Method and apparatus for establishing network connection for a processor without an operating system boot|
|US6292569||Oct 4, 2000||Sep 18, 2001||Intertrust Technologies Corp.||Systems and methods using cryptography to protect secure computing environments|
|US6327652||Jan 8, 1999||Dec 4, 2001||Microsoft Corporation||Loading and identifying a digital rights management operating system|
|US6330588||Dec 21, 1998||Dec 11, 2001||Philips Electronics North America Corporation||Verification of software agents and agent activities|
|US6338139||Jul 24, 1998||Jan 8, 2002||Kabushiki Kaisha Toshiba||Information reproducing apparatus, authenticating apparatus, and information processing system|
|US6341373||Dec 1, 2000||Jan 22, 2002||Liberate Technologies||Secure data downloading, recovery and upgrading|
|US6363486||Jun 5, 1998||Mar 26, 2002||Intel Corporation||Method of controlling usage of software components|
|US6363488||Jun 7, 1999||Mar 26, 2002||Intertrust Technologies Corp.||Systems and methods for secure transaction management and electronic rights protection|
|US6367012||Dec 6, 1996||Apr 2, 2002||Microsoft Corporation||Embedding certifications in executable files for network transmission|
|US6378072 *||Feb 3, 1998||Apr 23, 2002||Compaq Computer Corporation||Cryptographic system|
|US6381741||May 18, 1998||Apr 30, 2002||Liberate Technologies||Secure data downloading, recovery and upgrading|
|US6389402||Jun 9, 1999||May 14, 2002||Intertrust Technologies Corp.||Systems and methods for secure transaction management and electronic rights protection|
|US6389537||Apr 23, 1999||May 14, 2002||Intel Corporation||Platform and method for assuring integrity of trusted agent communications|
|US6427140||Sep 3, 1999||Jul 30, 2002||Intertrust Technologies Corp.||Systems and methods for secure transaction management and electronic rights protection|
|US6449367||Feb 23, 2001||Sep 10, 2002||Intertrust Technologies Corp.||Steganographic techniques for securely delivering electronic digital rights management control information over insecure communication channels|
|US6477252 *||Aug 29, 1999||Nov 5, 2002||Intel Corporation||Digital video content transmission ciphering and deciphering method and apparatus|
|US6477648||Mar 23, 1997||Nov 5, 2002||Novell, Inc.||Trusted workstation in a networked client/server computing system|
|US6480948 *||Jun 24, 1999||Nov 12, 2002||Cirrus Logic, Inc.||Configurable system memory map|
|US6480961||Mar 2, 1999||Nov 12, 2002||Audible, Inc.||Secure streaming of digital audio/visual content|
|US6546489 *||Mar 4, 1999||Apr 8, 2003||Western Digital Ventures, Inc.||Disk drive which provides a secure boot of a host computer system from a protected area of a disk|
|US6557104 *||May 2, 1997||Apr 29, 2003||Phoenix Technologies Ltd.||Method and apparatus for secure processing of cryptographic keys|
|US6560706||Jan 21, 1999||May 6, 2003||Intel Corporation||Interface for ensuring system boot image integrity and authenticity|
|US20010025281||Feb 21, 2001||Sep 27, 2001||International Business Machines Corporation||Method for access control of aggregated data|
|US20020007452||Aug 11, 1997||Jan 17, 2002||Chandler Brendan Stanton Traw||Content protection for digital transmission systems|
|US20020069395||Nov 29, 2001||Jun 6, 2002||Hiroaki Fujiwara||Method and apparatus of checking mount quality of circuit board|
|US20020107803||Aug 23, 2001||Aug 8, 2002||International Business Machines Corporation||Method and system of preventing unauthorized rerecording of multimedia content|
|US20020120936||Mar 11, 2002||Aug 29, 2002||Del Beccaro David J.||System and method for receiving broadcast audio/video works and for enabling a consumer to purchase the received audio/video works|
|US20020152173||Apr 5, 2002||Oct 17, 2002||Rudd James M.||System and methods for managing the distribution of electronic content|
|1||"Facing an Internet Security Minefield,Microsoft Hardens NT Server Defenses", Young R., Windows Watcher, Sep. 12, 1997, vol. 7, Issue 9, p1, 6p, 1 chart.|
|2||"Internet Security: SanDisk Products and New Microsoft Technology Provide Copy Protected Music for Internet Music Player Market. (Product Annoucement)", Edge: Work Group Computing Report, Apr. 19, 1999, 2 pages.|
|3||"Phoenix Technologies Partners with Secure Computing in Enterprise Security Marketplace", Jul. 12, 2001, Business Wire, Courtesy of Dialog Text Search, p. 1-2.|
|4||"Secure Management Information Exchange", Bhatti et al., Journal of Network and Systems Management, vol. 4, No. 3, 1996, pp. 251-277.|
|5||"Security Agility for Dynamic Execution Environments", Petkac et al., IEEE 1999, pp. 377-390.|
|6||"Security Enhancement for the "Simple Authentication Key Agreement Algorithm"", Lin et al., IEEE 2000, pp. 113-115.|
|7||"Techniques for secure execution of mobile code: a review", J.M. Mas-Ribes and Benoit Macq, Ann. Telecommun., 55, n 7-8, 2000, pp. 1-9.|
|8||*||486 Microprocessors, SSV Software Systems PC/104 Products, http://www.ssv-embedded.de/ssv/pc104/p71.htm, Mar. 13, 1998.|
|9||Abadi et al., "Authentication and Delegation with Smart-cards", Jul. 30, 1992, 30 pgs.|
|10||Arbaugh et al., "A Secure and Reliable Bootstrap Architecture", Distributed Systems Laboratory, Philadelphia, PA, 1997, pp. 65-71.|
|11||Clark et al., "Bits: A Smartcard Protected Operating System", Communications on the ACM, vol. 37, No. 11, Nov. 1994, pp. 66-70, 94.|
|12||Lampson et al., "Authentication in Distributed Systems: Theory and Practice", Digital Equipment Corporation, ACM Transactions on Computer Systems, vol. 10, No. 4, Nov. 1992, pp. 265-310.|
|13||Murphy et al., "Preventing Piracy: Authorization Software May Ease Hollywood's Fear of the Net", Internet World Magazine, Apr. 1, 2000, 3 pages.|
|14||Schneier, B., "Applied Cryptography", Applied Cryptography. Protocols, Algoriths, and Source Code in C, 1996, pp. 574-577.|
|15||Yee, "Using Secure Coprocessors", School of Computer Science, Carnegie Mellon University, 1994, 104 pgs.|
|Citing Patent||Filing date||Publication date||Applicant||Title|
|US7093115 *||Dec 19, 2002||Aug 15, 2006||Intel Corporation||Method and apparatus for detecting an interruption in memory initialization|
|US7159240 *||Nov 16, 2001||Jan 2, 2007||Microsoft Corporation||Operating system upgrades in a trusted operating system environment|
|US7210009 *||Sep 4, 2003||Apr 24, 2007||Advanced Micro Devices, Inc.||Computer system employing a trusted execution environment including a memory controller configured to clear memory|
|US7216369 *||Jun 28, 2002||May 8, 2007||Intel Corporation||Trusted platform apparatus, system, and method|
|US7272832 *||Oct 25, 2001||Sep 18, 2007||Hewlett-Packard Development Company, L.P.||Method of protecting user process data in a secure platform inaccessible to the operating system and other tasks on top of the secure platform|
|US7313705 *||Dec 31, 2002||Dec 25, 2007||Texas Instrument Incorporated||Implementation of a secure computing environment by using a secure bootloader, shadow memory, and protected memory|
|US7392541 *||Jan 15, 2004||Jun 24, 2008||Vir2Us, Inc.||Computer system architecture and method providing operating-system independent virus-, hacker-, and cyber-terror-immune processing environments|
|US7496958 *||Oct 29, 2003||Feb 24, 2009||Qualcomm Incorporated||System for selectively enabling operating modes of a device|
|US7530103 *||Aug 7, 2003||May 5, 2009||Microsoft Corporation||Projection of trustworthiness from a trusted environment to an untrusted environment|
|US7600256 *||Feb 4, 2005||Oct 6, 2009||Microsoft Corporation||Security critical data containers|
|US7631196 *||Feb 25, 2002||Dec 8, 2009||Intel Corporation||Method and apparatus for loading a trustable operating system|
|US7676665 *||Sep 14, 2005||Mar 9, 2010||Intel Corporation||Arrangements for initialization-time and run-time integration of firmware and software extensions for supporting add-in hardware|
|US7725740 *||May 19, 2004||May 25, 2010||Nagravision S.A.||Generating a root key for decryption of a transmission key allowing secure communications|
|US7757098 *||Jun 27, 2006||Jul 13, 2010||Intel Corporation||Method and apparatus for verifying authenticity of initial boot code|
|US7765374 *||Jan 25, 2007||Jul 27, 2010||Microsoft Corporation||Protecting operating-system resources|
|US7765392||Jun 29, 2006||Jul 27, 2010||Intel Corporation||Method and apparatus for establishing processor as core root of trust for measurement|
|US7831839||Feb 3, 2006||Nov 9, 2010||Sony Computer Entertainment Inc.||Methods and apparatus for providing a secure booting sequence in a processor|
|US7882221 *||Jun 2, 2008||Feb 1, 2011||International Business Machines Corporation||Method and system for measuring status and state of remotely executing programs|
|US8001377||Jan 13, 2009||Aug 16, 2011||Sony Computer Entertainment Inc.||Methods and apparatus for secure data processing and transmission|
|US8090662 *||Aug 14, 2006||Jan 3, 2012||Napster, Llc||Method and apparatus for dynamic renewability of content|
|US8181040 *||Feb 6, 2009||May 15, 2012||Panasonic Corporation||Program execution device|
|US8185748||Feb 3, 2006||May 22, 2012||Sony Computer Entertainment Inc.||Methods and apparatus for facilitating a secure processor functional transition|
|US8201240 *||Sep 6, 2006||Jun 12, 2012||Nokia Corporation||Simple scalable and configurable secure boot for trusted mobile phones|
|US8250374||May 20, 2010||Aug 21, 2012||Intel Corporation||Method and apparatus for verifying authenticity of initial boot code|
|US8380987||Jan 25, 2007||Feb 19, 2013||Microsoft Corporation||Protection agents and privilege modes|
|US8386788 *||Nov 10, 2009||Feb 26, 2013||Intel Corporation||Method and apparatus for loading a trustable operating system|
|US8407476 *||Nov 10, 2009||Mar 26, 2013||Intel Corporation||Method and apparatus for loading a trustable operating system|
|US8463968 *||Mar 31, 2005||Jun 11, 2013||Intel Corporation||Mechanism for a shared serial peripheral interface|
|US8484232 *||Nov 22, 2005||Jul 9, 2013||International Business Machines Corporation||Method, computer arrangement, computer program and computer program product for checking for the presence of control statements in a data value|
|US8522053||Apr 6, 2012||Aug 27, 2013||Panasonic Corporation||Program execution device|
|US8745754 *||Sep 22, 2004||Jun 3, 2014||Scm Microsystems Gmbh||Device for secure access to digital media contents, virtual multi-interface driver and system for secure access to digital media contents|
|US8782434||Jul 15, 2011||Jul 15, 2014||The Research Foundation For The State University Of New York||System and method for validating program execution at run-time|
|US8782435||Jul 15, 2011||Jul 15, 2014||The Research Foundation For The State University Of New York||System and method for validating program execution at run-time using control flow signatures|
|US8874938||Jul 26, 2013||Oct 28, 2014||Panasonic Intellectual Property Corporation Of America||Program execution device|
|US8904189 *||Jul 15, 2011||Dec 2, 2014||The Research Foundation For The State University Of New York||System and method for validating program execution at run-time using control flow signatures|
|US8938796||Sep 13, 2013||Jan 20, 2015||Paul Case, SR.||Case secure computer architecture|
|US9122633||Jan 13, 2015||Sep 1, 2015||Paul Case, SR.||Case secure computer architecture|
|US9141829||Nov 27, 2013||Sep 22, 2015||Socionext Inc.||Secure processor and a program for a secure processor|
|US9218485||Sep 11, 2014||Dec 22, 2015||Panasonic Intellectual Property Corporation Of America||Program execution device|
|US9524404||Nov 16, 2015||Dec 20, 2016||Panasonic Intellectual Property Corporation Of America||Program execution device|
|US9536110||Nov 27, 2013||Jan 3, 2017||Socionext Inc.||Secure processor and a program for a secure processor|
|US20030097578 *||Nov 16, 2001||May 22, 2003||Paul England||Operating system upgrades in a trusted operating system environment|
|US20030101322 *||Oct 25, 2001||May 29, 2003||Gardner Robert D.||Protection of user process data in a secure platform architecture|
|US20030140238 *||Dec 31, 2002||Jul 24, 2003||Texas Instruments Incorporated||Implementation of a secure computing environment by using a secure bootloader, shadow memory, and protected memory|
|US20030163723 *||Feb 25, 2002||Aug 28, 2003||Kozuch Michael A.||Method and apparatus for loading a trustable operating system|
|US20040003288 *||Jun 28, 2002||Jan 1, 2004||Intel Corporation||Trusted platform apparatus, system, and method|
|US20040123088 *||Dec 19, 2002||Jun 24, 2004||Poisner David I.||Method and apparatus for detecting an interruption in memory initialization|
|US20040236874 *||Jan 15, 2004||Nov 25, 2004||Kenneth Largman||Computer system architecture and method providing operating-system independent virus-, hacker-, and cyber-terror-immune processing environments|
|US20040236959 *||May 19, 2004||Nov 25, 2004||Henri Kudelski||Security key generation method|
|US20050033980 *||Aug 7, 2003||Feb 10, 2005||Willman Bryan Mark||Projection of trustworthiness from a trusted environment to an untrusted environment|
|US20050055524 *||Sep 4, 2003||Mar 10, 2005||Advanced Micro Devices, Inc.||Computer system employing a trusted execution environment including a memory controller configured to clear memory|
|US20050097345 *||Oct 29, 2003||May 5, 2005||Kelley Brian H.||System for selectively enabling operating modes of a device|
|US20060020936 *||Sep 14, 2005||Jan 26, 2006||Wyatt David A||Arrangements for initialization-time and run-time integration of firmware and software extensions for supporting add-in hardware|
|US20060101310 *||Oct 22, 2004||May 11, 2006||Nimrod Diamant||Device, system and method for verifying integrity of software programs|
|US20060177068 *||Feb 3, 2006||Aug 10, 2006||Sony Computer Entertainment Inc.||Methods and apparatus for facilitating a secure processor functional transition|
|US20060179302 *||Feb 3, 2006||Aug 10, 2006||Sony Computer Entertainment Inc.||Methods and apparatus for providing a secure booting sequence in a processor|
|US20060179324 *||Feb 3, 2006||Aug 10, 2006||Sony Computer Entertainment Inc.||Methods and apparatus for facilitating a secure session between a processor and an external device|
|US20060179482 *||Feb 4, 2005||Aug 10, 2006||Microsoft Corporation||Security critical data containers|
|US20060224803 *||Mar 31, 2005||Oct 5, 2006||Amir Zinaty||Mechanism for a shared serial peripheral interface|
|US20060294020 *||Aug 14, 2006||Dec 28, 2006||Duet General Partnership||Method and apparatus for dynamic renewability of content|
|US20070067617 *||Sep 6, 2006||Mar 22, 2007||Nokia Corporation||Simple scalable and configurable secure boot for trusted mobile phones|
|US20070130620 *||Nov 22, 2005||Jun 7, 2007||International Business Machines Corporation||Method, computer arrangement, computer program and computer program product for checking for the presence of control statements in a data value|
|US20080005574 *||Jun 29, 2006||Jan 3, 2008||Cheng Antonio S||Method and apparatus for establishing prosessor as core root of trust for measurement|
|US20080022108 *||Jun 27, 2006||Jan 24, 2008||Brannock Kirk D||Method and apparatus for verifying authenticity of initial boot code|
|US20080183996 *||Jan 25, 2007||Jul 31, 2008||Microsoft Corporation||Protecting Operating-System Resources|
|US20080184373 *||Jan 25, 2007||Jul 31, 2008||Microsoft Corporation||Protection Agents and Privilege Modes|
|US20080209571 *||Sep 22, 2004||Aug 28, 2008||Scm Microsystems Gmbh||Device for Secure Access to Digital Media Contents, Virtual Multi-Interface Driver and System for Secure Access to Digital Media Contents|
|US20080235372 *||Jun 2, 2008||Sep 25, 2008||Reiner Sailer||Method and system for measuring status and state of remotely executing programs|
|US20090007100 *||Jun 28, 2007||Jan 1, 2009||Microsoft Corporation||Suspending a Running Operating System to Enable Security Scanning|
|US20090125717 *||Jan 13, 2009||May 14, 2009||Sony Computer Entertainment Inc.||Methods and Apparatus for Secure Data Processing and Transmission|
|US20090136038 *||Nov 27, 2007||May 28, 2009||Ememory Technology Inc.||Apparatus for receiving encrypted digital data and cryptographic key storage unit thereof|
|US20090150685 *||Feb 6, 2009||Jun 11, 2009||Hideki Matsushima||Program execution device|
|US20100058075 *||Nov 10, 2009||Mar 4, 2010||Kozuch Michael A||Method and apparatus for loading a trustable operating system|
|US20100058076 *||Nov 10, 2009||Mar 4, 2010||Kozuch Michael A||Method and apparatus for loading a trustable operating system|
|US20160056960 *||Sep 24, 2015||Feb 25, 2016||Intel Corporation||System and method for execution of a secured environment initialization instruction|
|U.S. Classification||713/193, 380/277, 713/100|
|International Classification||G06F1/24, G06F9/445|
|Cooperative Classification||G06F21/575, G06F9/4405, G06F15/177|
|European Classification||G06F9/44A2, G06F21/57B, G06F15/177|
|Apr 6, 2001||AS||Assignment|
Owner name: MICROSOFT CORPORATION, WASHINGTON
Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNORS:ENGLAND, PAUL;WILLMAN, BRYAN;REEL/FRAME:011707/0031;SIGNING DATES FROM 20010215 TO 20010219
|Jan 28, 2009||FPAY||Fee payment|
Year of fee payment: 4
|Oct 4, 2012||FPAY||Fee payment|
Year of fee payment: 8
|Dec 9, 2014||AS||Assignment|
Owner name: MICROSOFT TECHNOLOGY LICENSING, LLC, WASHINGTON
Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNOR:MICROSOFT CORPORATION;REEL/FRAME:034541/0001
Effective date: 20141014
|Apr 7, 2017||REMI||Maintenance fee reminder mailed|